PDF Action Pipeline (Batch Recipes)

Chain PDF operations into a reusable recipe

1Choose file2Settings3Process4Save

Choose a PDF and its contents will be analyzed and shown here.

Files are processed only in this browser. Nothing is uploaded to a server.

Chain several PDF operations into a recipe and run them in order

Continue in another tool
Task settings

Add the steps to run in order below and adjust their options. Save with a name to load and reuse the same recipe on your next visit. Each step runs the processing logic of a tool that already exists on this site, in order. Add several PDFs and the same recipe runs on each file in turn, with the results in a ZIP.

No steps added yet. Pick a step above and add it.

    Remove the password from an encrypted PDF first. The whole file is rewritten, so any existing digital signature becomes invalid. Keep the original and check the result.

    About PDF Action Pipeline (Batch Recipes)

    Chain any of these steps in order and save the chain as a named recipe: sanitize, grayscale, hairline fix, rich black to K100, remove hidden OCR text, watermark, page numbers, compress and PDF/A conversion. Each step calls the same processing code as the matching tool on this site, so it has the same limits.

    Recipes live in this browser's IndexedDB: they do not sync to other devices and disappear when site data is cleared. If any step fails, the run stops there and tells you which step failed and why.

    Orders that change the result

    Every step works on the previous step's output, so the order is the result. The usual mistakes are below; compress or sanitize after PDF/A, a semi-transparent watermark before PDF/A-1b and K100 after grayscale are also flagged as warnings under the step list while you build the recipe (the run isn't blocked).

    • Put PDF/A conversion last. A semi-transparent watermark added afterwards breaks PDF/A-1, which forbids transparency, and a sanitize step afterwards strips the XMP metadata that identifies the file as PDF/A.
    • Rich black to K100 after grayscale has nothing to do: grayscale already turned every color into gray. For print shops, choose one or the other.
    • Compress after watermarks and page numbers but before PDF/A, so duplicate resources from the new elements are cleaned up in the same pass.

    Behavior hidden in the default options

    The compress step's recommended preset downsamples images to 150 DPI at JPEG quality 0.75 and also removes metadata. That suits on-screen distribution; raise it to 300 DPI for print. Because metadata removal is on, compressing after PDF/A strips the PDF/A identification for the same reason sanitize does.

    The per-step sizes in the log show which step made the file bigger. Steps that embed a font, such as a watermark or page numbers in non-Latin text, add size, so compression should come after them.

    Worked examples

    An archive submission recipe

    Input
    Remove hidden OCR text → Sanitize → Watermark (opacity 0.2) → Compress → PDF/A-2b
    Output
    All 5 steps succeed; final file ready

    The watermark comes before PDF/A, and 2b allows transparency. With PDF/A-1b, or with sanitize moved to the end, the result may not meet archive requirements. See the guide "PDF/A and PDF/X Explained" for how the levels differ.

    How to use it

    1. Add steps, reorder them and adjust their options.
    2. Optionally save the chain as a named recipe.
    3. Load one PDF, run it, check the per-step log and download the result.

    Frequently asked questions

    Can I run several files at once?

    Yes. Add several PDFs and the same recipe runs on each file in turn; the results come as a ZIP of “originalname-pipeline.pdf” files. The log lists each step's result with its file name.

    If a step fails, can I get the partial result?

    No. The run stops and nothing is downloaded. With several files, the run also stops at the first file that fails, and the error names that file. Run that step on its own tool page to see the detailed reason.

    Can the same step appear twice?

    Yes, for example two watermarks with different text. Compressing twice, though, rarely shrinks already downsampled images further and can lower quality.

    Sources

    Last reviewed:

    Privacy and how data is handled